Instructor: | A. Erdem Sariyuce (erdem AT buffalo.edu) |

Class hours: | Wed 10:00-12:40, Greiner 134C/135C |

Office hours: | Wed 3:30-5:30, Online over Zoom (link) |

Graphs are everywhere. Their scale, rate of change, and the irregular nature pose many new challenges. Deep learning has been shown to be successful in a number of domains, ranging from images to natural language processing. However, applying deep learning to the ubiquitous graph data is non-trivial because of the unique characteristics. This seminar course covers recent papers in the last few years about deep learning on graphs. We will consider graph embeddings, knowledge graphs, graph kernels, graph neural networks, graph convolutional networks, graph adversarial methods. Students will learn the literature on deep learning on graphs, understand the state-of-the-art algorithms on various problems, and be familiar with the recent trends.

It is assumed that students have a solid background on discrete mathematics and algorithms. Basic research skills like paper reading, critical thinking, problem solving, report writing, communication, and presentation are important as well.

- Paper presentation: 20 pts
- Piazza questions: 7.5 pts (one question)
- Class participation: 10 pts (in total)
- Attendance: 3 pts (one class)

The final grade is S/U and 70 pts score is needed for an S.

Students will do the presentations. Tentative schedule is below. A presentation is expected to be 45 mins long. Each week, all the students will read the paper of the week before class and half of the students (I'll tell which half) will ask **a unique question** on Piazza (except the presenter) to facilitate a class discussion. Questions should be open-ended and provide ground for class discussions, i.e., 'can you explain alg 1?' is not that kind of question. Questions should be posted to Piazza by Monday night, 11.59 pm EST.

- Feb 2:
**Introduction, Course Overview by instructor****[Slides for the first class]**

- Feb 9:

**The Why, How, and When of Representations for Complex Systems SIAM Review 63(3), pp. 435-485, 2021***by Erdem*

- Feb 16:

**Geometric deep learning: going beyond Euclidean data IEEE Signal Processing Magazine 2017***by Jason*

- Feb 23:

**node2vec: Scalable Feature Learning for Networks KDD 2016***by Shubhangi*

- Mar 2:

**Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec WSDM 2018***by Mihir*

- Mar 9:

**The impossibility of low-rank representations for triangle-rich complex networks PNAS 2020***by Nikita*

- Mar 16:

**Complex Embeddings for Simple Link Prediction ICML 2016***by Ashley*

- Mar 23:

**NO CLASS (SPRING RECESS)**

- Mar 30:

**Semi-supervised classification with graph convolutional networks ICLR 2017***by Raksha*

- Apr 6:

**Graph attention networks. ICLR 2018***by Mahesh Bhosale*

- Apr 13:

**Inductive Representation Learning on Large Graphs NIPS 2017***by Naveen*

- Apr 20:

**GNNExplainer: Generating Explanations for Graph Neural Networks NeurIPS 2019***by Sanjay Krishnaan*

- Apr 27:

**Weisfeiler and Leman go neural: Higher-order graph neural networks. AAAI 2019***by Siddhesh*

- May 4:

**Combining Label Propagation and Simple Models Out-performs Graph Neural Networks. ICLR 2021***by Deepak*

- May 11:

**Adversarial Attack on Graph Structured Data ICML 2018***by Jakir*